Papers by Julien Velcin
HISTOIRESMORALES: A French Dataset for Assessing Moral Alignment (2025.naacl-long)
Copied to clipboard
Thibaud Leteno, Irina Proskurina, Antoine Gourru, Julien Velcin, Charlotte Laclau, Guillaume Metzler, Christophe Gravier
| Challenge: | HistoiresMorales is a dataset based on moralStories in French . it is based upon annotations of moral values within the dataset . |
| Approach: | They propose a dataset in French that aims to align language models with moral values . they use annotations to ensure their alignment with French norms . |
| Outcome: | The proposed dataset guarantees grammatical accuracy and adaptation to the French cultural context. |
Towards Multi-Document Question Answering in Scientific Literature: Pipeline, Dataset, and Evaluation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing QA systems do not strictly enforce cross-document synthesis or exploit the explicit inter-paper structure that links sources. |
| Approach: | They propose a pipeline methodology for constructing a multi-document academic QA dataset . they detect communities based on citation networks and leverage Large Language Models . |
| Outcome: | The proposed method generates QA pairs related to multi-document content automatically and forms coherent communities based on citation networks and large language models. |
Monitoring geometrical properties of word embeddings for detecting the emergence of new topics. (2021.emnlp-main)
Copied to clipboard
| Challenge: | a recent study shows that slow emerging topics are often detected too late . a positive correlation is linked with event-like topics while a negative correlation is a sign of emergence. |
| Approach: | They propose to monitor words representation in embedding space and use one of its geometrical properties to characterize the emergence of topics. |
| Outcome: | The proposed method outperforms state-of-the-art methods on two public datasets of press and scientific articles. |
When Quantization Affects Confidence of Large Language Models? (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing studies have shown that quantization compromises performance and exacerbates biases in Large Language Models. |
| Approach: | They propose an explanation for quantization loss based on confidence levels . they propose a range of efficient compression and acceleration methods including quan-tization . |
| Outcome: | The proposed methods show that quantization decreases confidence regarding true labels and that it exacerbates biases across different scales. |
Structure-Aware Quantized Retrieval for Long-Document Question Answering (2026.findings-acl)
Copied to clipboard
| Challenge: | Traditional long-document QA/RAG pipelines suffer from context fragmentation . Graph Neural Networks (GNNs) can encode hierarchy, but deep GNNs tend to over-smooth representations. |
| Approach: | They propose a framework that aligns hierarchical graph representations with a universal token vocabulary and integrates explicit structure into retrieval. |
| Outcome: | The proposed framework captures universal hierarchical patterns rather than overfitting to specific layouts. |